How it works · CV Review

Turn your CV into a shortlist.

Calibrd reviews your CV like an ATS and a recruiter, then rewrites weak sections into stronger, interview-winning results — one before-and-after at a time.

✦ Free to start · part of your 8 AI credits
Already installed

Open the side panel from the Calibrd icon in your toolbar, switch to the CV tab, and hit Analyse. A review is 5 credits and every account starts with 8, so your first CV review and edits are free.

Add Calibrd to Chrome calibrd.com
Senior ML Engineer, Spam…▾
Researching
MC
Report
CV
Cover Letter
Practice
ReviewDIAGNOSE
EditAPPLY EDITS
✓
ml-engineer-cv-two-column.pdf
Run analysis below to get your ATS and recruiter scores.
⚠ Extraction issue detected · review after analysis
Two-column layout detected. We extracted your CV column-by-column, but the reading order may not match what you intended.
Run CV Review
✦ -5 credits138 remaining
Analyse
ReviewDIAGNOSE
EditAPPLY EDITS
48
ATS score
How ATS ranks your CV
52
Recruiter score
How fast a recruiter reads it
Solid ML foundation, but vague responsibility-dump bullets and missing trust-and-safety signals will likely filter this CV out before a human sees it.
01What the AI seesATS + JD parse
ATS Layout Two-ColumnAI screening+8 points
▾
No Quantified ImpactHuman · AI+10 points
▾
02What a human sees~38s skim
Weak Summary SignalHuman+7 points
▾
ReviewDIAGNOSE
EditAPPLY EDITS
01What the AI seesATS + JD parse
No Quantified ImpactHuman · AI+10 points▴
All bullets across both roles use vague language with zero metrics, percentages, or scale indicators.
Angle: add concrete numbers to at least two bullets per role — latency reduction %, dataset size, accuracy lift.
“Improved model performance and helped reduce latency across services. Utilized various tools and technologies to support and scale ML pipelines.”
Suggested rewrite
Improved recommendation model precision by [X]% and reduced p99 serving latency by [X]ms across services handling [X]M daily requests; scaled ML pipelines to support [X]TB of daily training data.
Preview change
ReviewDIAGNOSE
EditAPPLY EDITS
No Quantified ImpactHuman · AI+10 points
Before
Improved model performance and helped reduce latency across services. Utilized various tools and technologies to support and scale ML pipelines.
AfterEDIT
Improved recommendation model precision by [X]% and reduced p99 serving latency by [X]ms across services handling [X]M daily requests; scaled ML pipelines to support [X]TB of daily training data.
Confirm & apply
Cancel
ReviewDIAGNOSE
EditAPPLY EDITS
⌄TechCorp· 3✓ Tailored↑↓⋮
Company
TechCorp
Job title
Senior Machine Learning Engineer
Dates
2021 – Present
Diff vs master CV
Responsible for building and maintaining machine learning models for the recommendation system. Worked closely with cross-functional teams to deliver data-driven solutions. Improved model performance recommendation model precision by [X]% and helped reduce latency across services. Utilized various tools and technologies to support and scale ML pipelines. reduced p99 serving latency by [X]ms across services handling [X]M daily requests; scaled ML pipelines to support [X]TB of daily training data.
✦ Calibrate · 1↶ Revert
⚠3 major issues still unfixed — fix those first for the bigger gain.
Save to tailored
Save & update tailored score · -2 cr
ReviewDIAGNOSE
EditAPPLY EDITS
OriginalTailored
Maya Chen
Machine Learning Engineer · maya.chen@email.com · linkedin.com/in/mayachen
Summary
Machine Learning Engineer with 5+ years building impactful, scalable ML systems — recommendations, ranking, and trust-and-safety.
Experience
Senior ML Engineer2021 – Present
TechCorp
  • —Improved recommendation model precision by 24% and cut p99 latency 35% across services handling 8M daily requests.
  • —Built CI/CD + observability adopted by 12 squads.
  • —Led the trust-and-safety ranking workstream end to end.
Export
PDFPDF
Opens print → Save as PDF
DOCXWord
Coming soon
MDMarkdown
Plain text download
↗Copy to LinkedIn
Summary to clipboard
1In the CV tab, run the review — it flags a two-column ATS risk
The loop, step by step

Six steps. About five minutes, start to export.

✓ml-engineer-cv-two-column.pdf
⚠ Two-column layout detected — flagged before any credits are spent.
Analyse✦ -5 credits
01

Upload and run the review

Drop your CV into the CV tab. Calibrd checks the file before charging anything, so layout traps like two-column templates are flagged up front and you never spend credits on a noisy read.

48
ATS score
How ATS ranks it
52
Recruiter score
~38s skim
No Quantified Impact+10 points▾
02

Two scores, every gap ranked

You get an ATS pass and a recruiter skim, scored separately, because they fail you for different reasons. Each finding shows the points it's worth, so you fix the biggest wins first.

No Quantified ImpactHuman · AI+10 points
Angle: add concrete numbers to two bullets per role — latency %, dataset size, accuracy lift.
Suggested rewrite
Improved recommendation model precision by [X]% and reduced p99 serving latency by [X]ms across services handling [X]M daily requests.
Preview change
03

Open a gap, get the rewrite

Every finding comes with a concrete rewrite. Where a number belongs, you get an [X] placeholder to fill in yourself. Calibrd never invents figures about your work.

Before
Improved model performance and helped reduce latency…
After
Improved recommendation model precision by 24% and cut p99 latency 35%…
Confirm & apply
04

Approve it, before and after

Nothing changes without you. Put the original next to the rewrite, edit the wording if you want, then confirm and apply, one bullet at a time.

TechCorp· 3✓ Tailored
Diff vs master CV
Improved model performance recommendation model precision by [X]% and reduced p99 latency by [X]ms.
Save to tailoredSave & update tailored score · -2 cr
05

Calibrate any bullet, save to a tailored CV

The Edit tab is yours to refine: AI-calibrate any single bullet for 1 credit, or edit the wording by hand. Everything saves to a job-specific tailored CV — your master CV stays untouched, so each application keeps its own version.

Maya Chen
Machine Learning Engineer · single column · ATS-readable
PDFPDF
DOCXWord
MDMarkdown
↗LinkedIn
06

Export clean

Preview the rendered CV and export a single-column, ATS-readable PDF. A Markdown version and a copy-to-LinkedIn summary are there too; Word export is on the way.

See it for real

The whole loop, recorded in the actual product.

Thirty seconds, no editing tricks: upload, review, one rewrite applied, PDF out. Tap to watch it large.

Fair questions

Before you spend a credit.

No. Where a rewrite needs a figure it leaves an [X] placeholder, like “improved precision by [X]%”, for you to fill in. Nothing reaches your CV until you type the real number and approve the change.

Every account starts with 8 AI credits, so your first CV review and a round of edits are free. A review is 5 credits, re-scoring after edits is 2, and AI-calibrating a single bullet is 1. Every button shows its cost before you click, so nothing is ever spent by surprise.

No. Applied edits are saved to a tailored CV for that one job. Your master CV stays exactly as it is, and each application can keep its own tailored version.

Many ATS parsers read PDFs one column at a time, which scrambles the order you intended and can drop whole sections. Calibrd warns you before you spend a credit, and every export template is single-column and ATS-readable by design.

A clean, single-column PDF you save straight from the print dialog, a Markdown version for plain-text applications, and a short summary you can paste into LinkedIn. Word export is on the way.

Run your first review.

Calibrd icon in your toolbarCV tabAnalyse

5 credits · about a minute · every change is yours to approve

Don’t have it yet? Add Calibrd to Chrome
How CV Review works — Calibrd